Formation of nanocrystalline semiconductor particles within a bicontinuous cubic phase

Lead nitrate (Pb(NO 3 ) 2 ), sodium sulfide (Na 2 S), chloroform (HPLC grade) and n-dodecanethiol were obtained from Aldrich. AOT and methanol (99.9%) were purchased from Sigma and used as received. Water was purified by B-pure system (Barnstead) providing 17 megohm-cm quality at outlet.

The n-dodecanethiol capped nanocrystallites were synthesized typically using the following procedure. About 1.5 g mixture of aqueous Na 2 S solution (21.5 wt %) and AOT (78.5 wt %) was prepared in a 5 ml vial followed by centrifugation for 2 minutes at 4000 rpm. After two hours, the mixture was stirred by using a spatula until visibly uniform. After 24 hour incubation at room temperature, the mixture became transparent and extremely viscous. The presence of the cubic phase was ascertained by observing the sample under a polarized optical microscope. (BCP is optically isotropic because of its cubic symmetry and therefore appears dark between crossed polarizers. On the other hand, the lamellar and hexagonal phases which exist on either side of BCP in the phase diagram (see FIG. 3 are highly birefringent)). For Na 2 S solutions with higher concentrations (above 0.10 M), slightly more than 21.5% of the solution was added into AOT in order to reach cubic phase. As shown in FIG. 2(a), the sample in cubic phase was then filled into 5 ml disposable syringe 20 (without needles) and extruded in the shape of a cylinder 22 of 2 mm in diameter. Injection of the cylinder into aqueous 0.40 M Pb(NO 3 ) 2 solution 24 resulted in homogeneous nucleation inside the BCP. For each experiment, the sample was made using freshly prepared Na 2 S solution. After typically 1 hour, cylinder 22 was taken out from Pb(NO 3 ) 2 solution and washed with deionized water and then dissolved in 10 ml methanol. Centrifugation for 3 minutes at 4000 rpm led to complete precipitation of dark brown PbS powder. The colorless methanol solution was discarded by decantation. The methanol wash followed by centrifugation and decantation was repeated once again. Methanol (10 ml) and an excess n-dodecanethiol (with respect to the molar number of sulfide) were then added to the sediment followed by vigorous shaking. After 10 minutes, dark PbS powder was recovered through centrifugation and purified by washing twice with methanol. Six different PbS nanoparticulate samples were prepared by starting with aqueous Na 2 S solution with concentrations of 0.025, 0.050, 0.070, 0.10, 0.20 and 0.30 M, respectively. The powder was then dissolved in chloroform giving clear yellow brown or dark brown solutions depending on concentration and particle size. UV-visible absorption spectra were taken on HP8452A photodiode array spectrophotometer.

Small-angle X-ray diffraction studies of the cubic phase of AOT mixed with Na 2 S solutions were performed at the National Synchrotron Light Source at Brookhaven National Laboratory using 8 keV photons and a Fuji image plate system. The AOT/Na 2 S samples extruded from 1 ml syringe (without needle) were filled into 0.7-mm-diameter capillaries and sealed to prevent loss of water due to evaporation.

Powder X-ray diffraction spectra were collected on a Rigaku two-circle diffractometer using CuKα--radiation from a Rigaku 200 kV rotating anode X-ray generator. Sample for X-ray diffraction was prepared by drying the chloroform solution on a clean silicon wafer. The particle size of PbS samples was determined by measuring the full width at half maximum (FWHM) of the observed X-ray diffraction peaks. A diffraction scan taken for a NaCl sample of known grain size (>100 nm) was used to determine the peak broadening arising due to instrumental effects such as the divergence of the incident beam and the width of the X-ray source. As peak broadening due to particle size is only observable for grains less than 100 nm in diameter, a measurement of peak width in the larger grained standard sample allows the aforementioned effects to be segregated. Using the method described in B. D. Cullity, Elements of X-ray Diffraction (Addison-Wesley, Reading, Mass. (1978) p284 (the entirety of which is incorporated herein by reference for all purposes), which assumes that the diffraction line has the shape of an error curve, B the broadening of the diffraction peak arising due to particle size (radians) is given by

B.sup.2 =B.sub.M.sup.2 -B.sub.S.sup.2 (1)

where B M is the measured FWHM from the sample and B S is the FWHM of the diffraction peak from the standard sample.

Transmission election micrographs were taken on Hitachi-8100 operating at 200 kV. Carbon-coated formvar grids (300 mesh, from Ted Pella) were used as substrates for TEM characterizations. Langmuir-Blodgett technique was used to prepare monolayers of PbS crystallites for TEM studies. A few drops of thiol capped PbS nanoparticles in chloroform were spread on the surface of deionized water. Patches of thin films formed on the surface of water, without the application of surface pressure, were transferred onto formvar coated copper grids.

The experimental setup for a qualitative comparison of the mobilities of lead cations and sulfide anions within cubic AOT system is shown in FIG. 2(b). Glass capillary (diameter=1.5 mm) 30 was filled with mixture 32 of AOT (78 wt %) and water (22 wt %) exhibiting BCP. Immediately after the upper part of hollow capillary 30 was filled up with 0.20 M Na 2 S solution 34 (no air between AOT mixture/Na 2 S interface), the lower end of capillary 30 was submerged into 0.10 M Pb(NO 3 ) 2 solution 36. After about 15 minutes, formation of PbS was observed over region 38 (2 mm long) close to the top part of the sample, where the sample turned to light brown color. The formation of PbS was observed always close to the Na 2 S end, for all concentrations of Na 2 S and Pb(NO 3 ) 2 studied.

The stability of the cubic phase during the formation of PbS was studied under a polarizing optical microscope. A thin section of BCP (˜100 μm thick and 5 mm diameter) was sandwiched between a microscope slide and a coverslip. The free space around the sample was filled with Pb(NO 3 ) 2 solution, by capillary action. The changes in the appearance of the sample as the Pb 2+ ions diffused into it was observed periodically.

Phase diagram of AOT/Na 2 S/water. The phase diagram of binary AOT/water system shows that the cubic phase exists in a narrow region (18-25 wt % AOT) between inverse hexagonal phase (water poor side) and lamellar phase (water rich side). From small angle x-ray scattering (SAXS) studies combined with polarized optical microscopy is was determined that the cubic phase (Ia3d) still existed in ternary AOT/Na 2 S /water system. Increasing Na 2 S concentration led to the narrowing and shifting of cubic region to water-rich side and eventual disappearance when Na 2 S concentration was higher than 0.5M. This phase behavior is schematically shown by the phase diagram in FIG. 3. The phase boundaries in FIG. 3 are only estimations rather than representing precise values. An interesting feature observed in the low-angle X-ray diffraction (not shown) obtained from mixture of AOT and Na 2 S exhibiting BCP was the presence of diffraction spots instead of the usually observed powder diffraction rings. This indicated that the structure of the cubic phase in the capillary was coherent over a very long range. The samples used for PbS synthesis should have similar long range order (since they were prepared in exactly the same fashion), thus allowing uniform diffusion of lead cations throughout the sample.

UV-Visible Absorption Spectra. The absorption spectra of n-dodecanethiol capped PbS particles prepared from different concentrations of sodium sulfide solution mixed with AOT is shown in FIG. 4. The absorption in the region under 260 nm is due to the solvent (chloroform). As the concentration of Na 2 S solution incorporated in AOT increased from 0.025 M to 0.10 M, the shapes of the absorption spectra in the region 260-800 nm were almost the same. However, when the concentration of Na 2 S was increased to 0.20 M, a small shoulder appeared at around 300 nm. This shoulder became more pronounced and red-shifted to longer wavelength when the concentration of Na 2 S increased to 0.30 M or higher, indicating an increase in the absorption edge with size. All these curves were featureless in contrast to prior art absorption spectra of PbS nanocrystallites prepared from Pb(ClO 4 ) 2 in polyvinyl alcohol polymer matrix, which exhibited an excitonic peak around 600 nm. Since the size distribution of PbS particles prepared from Na 2 S solution with concentrations lower than 0.20 M was quite narrow, it is probably the surface structure of the PbS nanocrystallites which is responsible for the lack of excitonic features.

X-ray Diffraction Experiments. X-ray diffraction (XRD) of lead sulfide nanoparticles prepared from the exposure of Na 2 S solution incorporated in cubic phase of AOT to 0.40 M Pb(NO 3 ) 2 solution have shown that the concentration of Na 2 S solution had great effect on the size of PbS nanocrystallites which can be seen from FIG. 5. All the observed diffraction peaks for all particle sizes can be indexed on the basis of rocksalt structure. As the concentration of Na 2 S decreased from 0.30 M to 0.025 M, the width of the diffraction peaks increased as a result of the decrease in the size of the nanocrystallites. The average sizes, which are listed in Table 1, were calculated from the half width of the {111} and {200} peaks using the Scherrer formula with the correction due to instrument broadening as mentioned in experimental section.

______________________________________

The relationship between the concentration of aqueous Na.sub.2 S

solution in AOT forming cubic phase and the PbS nanocrystallites

size, calculated from the width of {111} and {200} peaks of XRD

experiments.

Na.sub.2 S, M

av. diameter, nm

______________________________________

0.025 4.0

0.050 4.7

0.070 5.8

0.10 6.6

0.20 8.2

0.30 13.5

______________________________________

Transmission Electron Microscopy. As shown by a transmission electron micrograph of lead sulfide nanocrystallites prepared from 0.050 M Na 2 S solution, the particles are slightly aspherical and tend to close pack into a quasi-hexagonal lattice. About half of the particles in the same micrograph gave sharp lattice fringes with most frequent spacings of 3.1 and 2.2 Å which correspond to (200) and (220) planes respectively. For particles which showed lattice fringes with good contrast, sizes could be measured by averaging the shorter and longer dimensions. The mean size obtained from the partial measurement in that micrograph was 4.9 nm±0.6 nm, which agrees very well with the size (4.7 nm) obtained from X-ray diffraction peaks (see Table 1). Precise measurement of particle size for about half of the particles was difficult due to poor contrast of the image. Therefore, the true standard deviation in the particle size may be better than that reported here. This is a significant improvement in the uniformity of size and morphology compared to that of PbS particles synthesized in polymer and sol-gel matrices, which yielded wide size distribution (50% deviation in the polymer) and various crystallite morphologies (both cubic and rod-like nanocrystallites were found when made by sol-gel method). The above-mentioned micrograph represented a typical appearance of the lead sulfide nanoparticulate samples prepared from Na 2 S solutions with concentration lower than 0.10M, except that the particle sizes decreased from 6 nm down to 4 nm as the concentration of Na 2 S dropped from 0.10 M to 0.025 M. When the concentration of Na 2 S solution was higher than 0.20 M, the particle size became much larger and a bimodal distribution in size was observed. A TEM micrograph of PbS particles prepared from 0.30 M Na 2 S within AOT matrix showed large particles with average size of about 25 nm and predominantly cubic morphology. In addition to these large particles, a large number of small particles (4-5 nm) with low contrast were visible

A high resolution TEM image of PbS nanocrystallites prepared from 0.10M Na 2 S showed lattice fringes with either 3.0 Å or 2.2 Å spacing corresponding to the (200) or the (220) planes, respectively. The reason that many PbS nanoparticles in the same area gave (200) and (220) fringes might be due to the cubic morphology of the particles which enables the particles to sit on its (001) face. The examination of those particles which gave sharp lattice fringes did not reveal lattice defects such as stacking faults. The lattice fringes extend to the edges of the particles indicating good crystallinity.

Experiments conducted by reversing the process, i.e., forming BCP of AOT in Pb(NO 3 ) 2 solution and exposing to Na 2 S solution, resulted in the formation of PbS only on the outer surface of the cylinder. This result is in conformity with the mobility studies (see FIG. 2(b)), described in the experimental section, which showed that the Pb 2+ cations diffuse much faster than the S 2- anions within cubic phase. This relative immobility of sulfide anions (S 2- ) with respect to lead cations (Pb 2+ ) within the cubic structure ensures uniform environment throughout the sample for nucleation and crystallization of PbS as the Pb 2+ ions diffuse from the surface into the core of the extruded cylinder. Presuming the density of AOT in cubic phase to be 1.2 g/cm 3 based on the fact that AOT in cubic phase is heavier than water (sinking instead of floating), it can be calculated that each pore consists of about 200 AOT molecules. In the mixture of 22 wt % 0.10 M Na 2 S and 78 wt % AOT, the ratio between number of AOT molecules and number of sulfide (S 2- ) is about 80:1, which means that each pore contains less than 3 sulfide anions. Thus, a probable process of PbS particle formation can be speculated as such: the infusion of large amount of Pb 2+ cations from the surrounding solution into the cylinder results in nucleation sites in some pores which subsequently attract sulfide anions from the pores in its vicinity leading to the growth of nanocrystallites. Another possibility which cannot be excluded is the simultaneous nucleation of PbS molecules and subsequent growth of the particles by Ostwald ripening.

Optical microscopic studies were carried out to ascertain that the substrate is in the cubic phase during the formation of PbS nanocrystallites. As mentioned earlier, BCP being optically isotropic, does not allow light to pass through when observed between crossed polarizers. However, in a slightly off-crossed state (obtained by rotating either the analyzer or the polarizer by a few degrees) a uniform orange color was observed along the periphery of the sample as soon as the Pb(NO 3 ) 2 solution was added. This reaction front moved into the sample at the rate of about 1-2 mm/hr. for all the samples. These studies enabled us to fix two experimental parameters during the synthesis of PbS nanoparticles. Firstly, it enabled us to ascertain that the BCP was maintained during the particle formation. Secondly, based on these observations, we could determine the period over which the substrate is exposed to Pb(NO 3 ) 2 solution. It should be mentioned here that for some combinations of Na 2 S and Pb(NO 3 ) 2 concentrations, it was observed that even though the reaction front was in the cubic phase, it left the substrate in the lamellar phase in its wake. It was found that as the difference in the concentrations of Na 2 S and Pb(NO 3 ) 2 solutions decreased, the lamellar phase region increased. This probably is the reason for the bimodal distribution observed for the 0.20 and 0.30 M concentrations of Na 2 S.

Lead sulfide nanocrystals by alternate method according to the present invention

The lead sulfide nanocrystals were also synthesized by first forming the cubic phase in a sulfide containing solution and then adding the one of the lead salt solutions. A mixture of surfactant (AOT) and sulfide containing solution (concentration variable 0.01-1M) was first prepared and then injected into lead compound solution in form of thin sheet or cylinder. After typically 1 hour, the solution was discarded by decantation. The viscous mixture was washed by deionized water followed by isolation process: dissolution in methanol, sonication and centrifuging at 4000 rpm rate. The obtained nanoparticles were purified by repeating dispersion in methanol, sonication and centrifuge cycle.

Cadmium sulfide nanocrystals

Cadmium sulfide nanoparticles were synthesized using one of the methods described in Example 1 and Example 1A, except the lead compound is replaced by cadmium compound (e.g. cadmium chloride, cadmium perchlorate, cadmium nitrate, eta.). FIG. 6 shows the absorption spectrum of obtained cadmium sulfide nanoparticles dispersed in chloroform.

lead selenide nanocrystals

Lead selenide nanoparticles were synthesized by same methods and processing procedures used in example 1 except that the sulfide-containing solution was replaced by a selenide-containing solution that was prepared by bubbling dilute aqueous sodium hydroxide solution with hydrogen selenide gas.

Mn-doped zinc sulfide nanocrystals

The synthesis was achieved by using the method used in EXAMPLE 1 except that the lead compound solution was replaced by a mixture of zinc chloride and trace amount manganese chloride (0.05-2%). When the surface of the Mn-doped ZnS particles was passivated and protected by thiol compounds, 25% of luminescent efficiency was achieved. FIG. 7 shows the photoluminescence (excitation and emission spectra) of mercaptopropionic-acid-capped ZnS:Mn nanoparticles dispersed in water. For comparison, the photoluminescence of micro-sized ZnS:Mn particulate sample is also shown.

Additional information concerning the present invention may be found in Yang et al., J. Phys. Chem., Vol. 100, No. 43, pages 17255-17259, the entirety of which is incorporated herein by reference.

Obviously, many modifications and variations of the present invention are possible in light of the above teachings. It is therefore to be understood that, within the scope of the appended claims, the invention may be practiced otherwise than as specifically described.